The Association Between Generative AI Use and Homogenization in Entrepreneurship Education: Manifestations, Potential Mechanisms, and Implications

Abstract

The rapid adoption of generative artificial intelligence tools in higher education raises important questions about innovation and entrepreneurship education. Through literature analysis, text mining, and classroom observation, this study systematically examines the homogenization phenomenon observed in undergraduate entrepreneurship course outcomes among AI users and explores potential mechanisms underlying this association. Using a mixed-methods approach based on 150 business plan samples and 45 in-depth interviews from three Chinese universities, the study finds that homogenization manifests primarily in four dimensions: clustering of project topics, templated business logic, converging data citations, and standardized language styles. Potential mechanisms underlying this association may include training data bias and information flattening at the technical level, shallow learning patterns and cognitive authority transfer at the cognitive level, and lagging assessment standards and curriculum design disconnection at the educational ecosystem level. Given the cross-sectional nature of this study, causal inference is limited, and alternative explanations including self-selection bias and temporal confounding cannot be ruled out. The research reveals patterns that raise concerns about the core objectives of entrepreneurship education in cultivating students' innovation capabilities and critical thinking, potentially indicating "skills hollowing-out" and loss of innovation ecosystem diversity among certain patterns of AI use. This study provides a theoretical framework and empirical evidence for understanding the transformation of entrepreneurship education in the AI era.